Apache OpenOffice (AOO) Bugzilla – Issue 80669
help: Crash when displaying help index items
Last modified: 2007-11-01 11:10:40 UTC
Autotest:framework\help\help.bas: The test opens the help, goes to the index page and tries to display each item of the index for each application and breaks at random positions. osl_releaseVolumeDeviceHandle /net/jumbo2.germany/ws/SRC680/src.m225/sal/osl/unx/file.cxx:2424 CallSystemHandler /so/ws/SRC680/src.m225/sal/osl/unx/signal.c:767 libc.so.1 + 0x9fa0f -- could not find checksum in database libc.so.1 + 0x9602f -- could not find checksum in database libc.so.1 + 0x961af -- could not find checksum in database com::sun::star::uno::Reference<com::sun::star::sdbc::XRow>ucbhelper::Content::getPropertyValuesInterface(const com::sun::star::uno::Sequence<rtl::OUString>&) /net/jumbo2.germany/ws/SRC680/src.m225/ucbhelper/source/client/content.cxx:755 com::sun::star::uno::Sequence<com::sun::star::uno::Any>ucbhelper::Content::getPropertyValues(const com::sun::star::uno::Sequence<rtl::OUString>&) /net/jumbo2.germany/ws/SRC680/src.m225/ucbhelper/source/client/content.cxx:703 com::sun::star::uno::Any ucbhelper::Content::getPropertyValue(const rtl::OUString&) /net/jumbo2.germany/ws/SRC680/src.m225/ucbhelper/source/client/content.cxx:653 unsigned char comphelper::MediaDescriptor::isStreamReadOnly()const /so/ws/SRC680/src.m225/comphelper/source/misc/mediadescriptor.cxx:0 ?? :0 ?? :0 ?? :0 _imp_setProcessLocale /net/jumbo2.germany/ws/SRC680/src.m225/sal/osl/unx/nlsupport.c:680 libc.so.1 + 0x9f609 -- could not find checksum in database libc.so.1 + 0x9f8f0 -- could not find checksum in database got frames from addr2line/database 5/5
Setting P2
Note: I have (at least) three different crash report ids which probably are the same problem: The issue IDs (internal) are: 150245, 150246 and probably another one the report-tool is currently fighting with
Setting Solaris (x86), currently verifying other platforms
Cannot reproduce on Linux/Patch as issue 80650 prevents the help from opening. Will try with FAT installation, just takes a little longer
On Windows there is no crash yet, but my machine is quickly running out of available memory, so we appear to have a memory leak. This is a patched m124 (StarOffice 8) to OOG680_m1.
The memory problem occurs first in m222. It's not sure that this is the reason for the crash but it's quite possible as the memory leak is rather big. ab->fme: As discussed...
Fixed in docsh.cxx Ready for QA.
.
The memory leak is fixed. I have still a few Server Timouts in the automated tests but they can be caused by to quick action on the dialog. The number of "crashes" has been reduced significantly (1/10) which is good. I set the issue to verified and nominate the CWS as soon as the minimum requirements for approval have been fulfilled. If there still is a problem we need to tackle that in another round.
*** Issue 80787 has been marked as a duplicate of this issue. ***
*** Issue 81060 has been marked as a duplicate of this issue. ***
*** Issue 79979 has been marked as a duplicate of this issue. ***
close